Chem. Mater.

1997, 9, 609-615

609

Formation of New Type of Porous Carbon by

Carbonization in Zeolite Nanochannels
Takashi Kyotani,* Takayuki Nagai, Sanjuro Inoue, and Akira Tomita
Institute for Chemical Reaction Science, Tohoku University, Katahira, Sendai 980-77, Japan
Received August 12, 1996. Revised Manuscript Received November 20, 1996X

An attempt was made to prepare porous carbon by using the channels of Y zeolite as a
template. Poly(acrylonitrile) and poly(furfuryl alcohol) were carbonized in the zeolite
channels and the resultant carbon/zeolite complexes were subjected to acid treatment in
order to extract carbon from the zeolite framework. In addition, pyrolytic carbon deposition
in the channels was carried out by exposing the zeolite to propylene at high temperature,
and then the carbon was liberated in the same manner as above. The morphology and
structure of the carbon prepared in the channels were characterized, and the results were
discussed in relation to the morphology and structure of the original zeolite template. It
was found that the microscopic morphology of the resultant carbons reflects that of the
corresponding zeolites. All of these carbons are highly porous, and some of the CVD carbons
have BET surface areas as high as >2000 m2/g.

Introduction
We have developed a template carbonization technique for synthesizing novel carbon materials. This
technique utilizes the inorganic substances whose openings or pores are controlled at nanometer level. Our
first attempt was to prepare thin graphite films by the
carbonization of organic polymers in the two-dimensional opening between the lamellae of a layered clay
such as montmorillonite and taeniolite.1-5 Recently, we
applied this technique to the preparation of onedimensional carbon such as carbon tubes by using an
anodic aluminum oxide film as one-dimensional template6,7 and obtained uniform hollow nanotubes with
open ends.
Zeolites have three-dimensional channels, and their
diamenter and shape are strictly controlled by the
framework topology. Because of the presence of such
channels, zeolites have been used as molecular-sieve
adsorbents or catalysts. In addition, the well-defined
nanospace in a zeolite has been utilized for the synthesis
of novel materials. If such a nanospace in a zeolite is
packed with carbon and then the carbon is extracted
from the zeolite framework, one can expect the formation of a porous carbon whose structure reflects the
porosity of the original zeolite template. On the basis
of this concept, we made a preliminary attempt to
carbonize poly(furfuryl alcohol) in zeolite channels and
found the formation of porous carbon.8 There were a
few other reports on carbon formation in zeolite channels. Bein et al. pyrolyzed poly(acrylonitrile) in the
channels of Y zeolite and mordenite and then liberated
X Abstract published in Advance ACS Abstracts, January 1, 1997.
(1) Kyotani, T.; Sonobe, N.; Tomita, A. Nature 1988, 331, 331.
(2) Sonobe, N.; Kyotani, T.; Hishiyama, Y.; Shiraishi, M.; Tomita,
A. J. Phys. Chem. 1988, 92, 7029.
(3) Sonobe, N.; Kyotani, T.; Tomita, A. Carbon 1991, 29, 61.
(4) Kyotani, T.; Yamada, H.; Sonobe, N.; Tomita, A. Carbon 1994,
32, 627.
(5) Kyotani, T.; Mori, T.; Tomita, A. Chem. Mater. 1994, 6, 2138.
(6) Kyotani, T.; Tsai, L.; Tomita, A. Chem. Mater. 1995, 7, 1427.
(7) Kyotani, T.; Tsai, L.; Tomita, A. Chem. Mater. 1996, 8, 2109.
(8) Kyotani, T.; Nagai, T.; Tomita, A. Extended Abstracts of Carbon
92, 1992, Essen, 437.

S0897-4756(96)00430-9 CCC: $14.00

the resultant carbon from the zeolites using HF solution.9 Since the carbon obtained thereby showed much
lower electronic conductivity than the carbon from bulk
poly(acrylonitrile), they concluded that the spatial limitation in the zeolite channels prevents the formation of
graphitized structure with high conductivity. Cordero
et al. prepared the carbon/zeolite complex with pyrolytic
carbon deposition from propylene over zeolite powder
as cracking catalyst, followed by dissolving the zeolite
with HF and HCl treatment to leave only carbon.10
Their main purpose was to observe the high-temperature oxidation behavior of the resultant carbon, in
comparison with that of pyrolytic carbon-deposited
carbon-carbon composites in the same conditions. Very
recently, they discussed the oxidation behavior of the
carbon extracted from zeolite in relation to its structure.11 Thus, except for our previous attempt,8 no one
has intentionally utilized zeolite as a template in order
to prepare porous carbon. Apart from zeolite, several
attempts were made to prepare porous carbon by the
template method using silica gel,12-14 porous glass,12,13
and sintered sodium chloride.15 None of these templates, however, has such regular channels as a zeolite
has.
Combined with the previous results, we report here
our attempt to prepare a new type of porous carbon by
using the channels of the Y zeolite as a template. The
morphology and structure of the carbon prepared in the
channels were characterized by several means. We
have paid special attention to how the morphology and
(9) Enzel, P.; Bein, T. Chem. Mater. 1992, 4, 819.
(10) Cordero, T.; Thrower, P. A.; Radovic, L. R. Carbon 1992, 30,
365.
(11) Rodrguez-Mirasol, J.; Cordero, T.; Radovic, L. R.; Rodrguez,
J. J., in preparation.
(12) Gilbert, M. T.; Knox, J. H.; Kaur, B. Chromatographia 1982,
16, 138.
(13) Knox, J. H.; Kaur, B.; Millward, G. R. J. Chromatogr. 1986, 3,
352.
(14) Kamegawa, K.; Yoshida, H. J. Bull. Chem. Soc. Jpn. 1990, 63,
3683.
(15) Pekela, R. W.; Hopper, R. W. J. Mater. Sci. 1987, 22, 1840.

1997 American Chemical Society

610

Chem. Mater., Vol. 9, No. 2, 1997

structure of the resultant carbon reflect those of the

zeolite.
Experimental Section
Zeolite Samples as a Template. The following three
types of Y zeolite were used in this study: USY(12) (H-form,
SiO2/Al2O3 g 12, Toso Inc., HSZ-360HUA), NaY(5.5) (Na-form,
SiO2/Al2O3 ) 5.5, supplied from Dr. H. Terasaki of Tohoku
University) and NaY(5.6) (Na-form, SiO2/Al2O3 ) 5.6, Catalysis
Society of Japan, JRC-Z-Y5.6). The first zeolite is an ultrastable one, which can keep its crystal structure even at 900
C. Though the others are not ultrastable, they can stand at
700 C. Among the three zeolites, NaY(5.5) has the highest
crystallinity.
Carbonization of Poly(acrylonitrile) and Poly(furfuryl alcohol). We attempted to carbonize poly(acrylonitrile)
(PAN) or poly(furfuryl alcohol) (PFA) in the zeolite channels
of NaY(5.5) and NaY(5.6). The introduction and polymerization of acrylonitrile in the channels were carried out in the
following way. After the powdery zeolite was dried under
vacuum at 150 C, acrylonitrile vapor at its saturated pressure
was allowed to contact with the dried zeolite at room temperature for 1 day. Then the acrylonitrile-zeolite complex was
evacuated for 30 min in order to remove acrylonitrile adsorbed
on the outer surface of zeolite, followed by -ray radiation (total
dose, 6 Mrad) to polymerize acrylonitrile.
In the case of furfuryl alcohol, it was introduced into NaY(5.6) in liquid phase. The dry powder of the zeolite was
impregnated with furfuryl alcohol under reduced pressure at
room temperature. The mixture of furfuryl alcohol and zeolite
powder was stirred for more than 5 days, and then it was
filtered, followed by washing with mesitylene to remove
furfuryl alcohol on the external surface of zeolite powder. The
polymerization of furfuryl alcohol in zeolite was carried out
by heating the powder under N2 flow at 80 C for 24 h and
then at 150 C for 8 h. The resultant PAN/zeolite and PFA/
zeolite complexes were heat-treated at a linear rate of 5 C/
min to 700 C and held there for 3 h to carbonize the polymers
in the zeolite channels. The carbon/zeolite complexes were
washed with an excess amount of 46% aqueous HF solution
at room temperature and subsequently refluxed in 36%
aqueous HCl solution to dissolve the zeolite framework. As a
result, carbon was obtained as an insoluble fraction.
Chemical Vapor Deposition (CVD) over Zeolite. Only
USY(12) was employed for CVD in zeolite channels, because
of its high thermal stability. The dried powdery zeolite was
placed in a vertical quartz reactor (i.d., 16 mm). The reactor
temperature was raised to 800 C under N2 flow, and then
propylene gas (2.5% in N2) was passed through the reactor at
a total flow rate of 200 cm3(STP)/min. The thermal decomposition of propylene in zeolite results in pyrolytic carbon
formation in its channels. After the desired period of time,
the reactor was cooled to room temperature. The zeolite
framework in the carbon/zeolite complex was dissolved by
successive washing with HF and HCl solutions, leaving only
carbon as an insoluble fraction in the solution.
Y zeolite has an open pore size of about 0.74 nm, which is
large enough for acrylonitrile, furfuryl alcohol and propylene
molecules to diffuse.
Characterization. The microscopic features of the complexes and the carbons were observed with a scanning electron
microscope (SEM, Hitachi, S900) and transmission electron
microscopes (TEM, JEOL: JEM-2000EXII, JEM-3010, and
JEM-ARM1250). To observe the inside of the sample, several
carbon particles were embedded in an epoxy resin and sliced
at a thickness of less than 100 nm with an ultramicrotome.
Then their TEM observation was performed. The structure
change of the complexes with the heat and acid treatments
and the crystallinity of the resultant carbons were examined
with an X-ray diffractometer (XRD, Shimadzu, XD-D1 with
Cu KR radiation) by referring to a silicon standard. The
surface area and the pore structure of the samples were
investigated with an automatic volumetric sorption analyzer
(Quantachrome, Autosorb-1) using N2 as adsorbent at -196
C. The surface area and the micropore volume were, respec-

Kyotani et al.
Table 1. Amount of Carbon in the Carbon/Zeolite
Complexes and the Results of Elemental Analysis for the
Carbons after the Acid Washing
carbon
precursor

zeolite

PAN
PAN
PFA
propylene
propylene
propylene
propylene

NaY(5.5)
NaY(5.6)
NaY(5.6)
USY(12)
USY(12
USY(12)
USY(12)

carbon
carbon
deposition content (g/g
period (h) of zeolite)

1
3
6
12

0.21
0.18
0.17
0.11
0.23
0.42
0.55

elemental analysis for

the carbon (wt %)
C

O (diff)

69
74
90
89
91
96
98

3
3
3
3
2
2
2

9
10
0
0
0
0
0

19
13
7
8
7
2
0

tively, determined from the BET equation and the DubininRadushkevich equation. The mesopore volume was calculated
by subtracting the micropore volume from the volume of N2
adsorbed at a relative pressure of 0.95.

Results
Elemental Analysis. Table 1 shows the amount of
the carbon formed in zeolites. This value was determined from the amount of the combustible part in each
carbon/zeolite complex. For all the complexes from PAN
and PFA, the amounts of carbon are almost 0.2 g/g of
zeolite. The carbon content in the complexes with
pyrolytic carbon increases with carbon deposition period.
The last part in Table 1 summarizes the results of the
elemental analysis for the carbon samples after the acid
washing. For all the samples, their ash content (not
shown here) was found to be less than 0.5%. Some of
the samples include significant amount of oxygen, even
though their precursors do not have oxygen, like PAN
or propylene. Most of such oxygen was probably incorporated into carbon during the washing process, because
the carbon/zeolite complexes from PAN and propylene
have almost no oxygen.
SEM Observation. Figure 1 shows the SEM photographs of the three types of zeolites, and their corresponding carbon/zeolite complexes and carbons. The
photographs of the zeolites (the left side of Figure 1)
confirm that the crystallinity of NaY(5.5) is highest
among the three zeolite samples. Some of the particles
seem to be a single crystal. The next highest one is
USY(12), whose particle size is somewhat smaller than
that of NaY(5.5). In both the zeolites, clear crystal faces
were observed on many particles. The particles of NaY(5.6) have a larger diameter (0.8-1.2 m), and they look
comprised of several small zeolite crystals. There is
almost no difference in morphology between the zeolites
and the corresponding complexes (the middle of Figure
1).
The right side of Figure 1 shows the SEM photographs of the carbon samples. The most remarkable
feature of the appearance of these carbons is a strong
resemblance to that of the corresponding template. For
instance, the PAN carbon particles from NaY(5.6)
appear to consist of several small parts. This feature
is exactly the same as in the corresponding template,
NaY(5.6). It was also the case with the PFA carbon
from NaY(5.6) (not shown here). Despite such resemblance, the particle size of the carbons from PAN or PFA
is always smaller than that of the corresponding zeolites. In the case of the CVD carbon, the smooth surface
of the carbon particles looks like a result of the inheritance of the crystal face of the USY(12) particles. The
particle size of these carbons is almost the same as that
of USY(12).

A New Type of Porous Carbon

Chem. Mater., Vol. 9, No. 2, 1997 611

Figure 1. SEM photographs of the zeolites (the left-hand side), the carbon/zeolite complexes (the middle) and the carbons (the
right): (a) NaY(5.5); (b) PAN carbon/NaY(5.5); (c) PAN carbon from NaY(5.5); (d) NaY(5.6); (e) PAN carbon/NaY(5.6); (f) PAN
carbon from NaY(5.6); (g) USY(12); (h) CVD carbon/USY(12) (CVD period, 3 h); (i) CVD carbon from the sample (h).

TEM Observation. Figure 2 presents the TEM

photographs of the carbons prepared from the polymers.
The bright-field image of the PAN carbon from NaY(5.5) (Figure 2a) displays a particle with a size of about
0.3 m, which corresponds to the carbon particles as
observed in the SEM photograph (Figure 1c). The
image is so dark that we cannot see its detailed

structure. The TEM image of the sliced sample of the

PAN carbon from NaY(5.5) is shown in Figure 2b. The
bright and gray parts in this image, respectively,
correspond to the cross sections of epoxy resin and
carbon particles embedded in the resin. Almost no voids
are observed in the carbon particles. In the bright-field
images of the PAN and PFA carbons from NaY(5.6)

612

Chem. Mater., Vol. 9, No. 2, 1997

Kyotani et al.

Figure 2. TEM images of the polymer carbons prepared in the zeolite channels (a microscope grid is seen under the samples):
(a, b) PAN carbon from NaY(5.5), ((b) is the TEM image of its sliced sample); (c) PAN carbon from NaY(5.6); (d) PFA carbon from
NaY(5.6).

(Figure 2c,d), dense and sparse portions are observed

and these images reflect the morphology which is seen
in the SEM photograph (Figure 1f).
Figure 3 shows the TEM photographs of the CVD
carbons with different carbon deposition periods. The
particles in the top TEM photograph correspond to the
carbon particles in Figure 1i. It should be noted that
the carbon particles with the 12 h deposition have thin
and denser carbon envelope on each particle, but the

other two images do not show such a layer. Since the

layer was observed only as the envelope of particles, it
must be ascribed to the carbon deposited on the external
surface of the zeolite particles.
XRD Analysis. The XRD patterns for the carbon/
zeolite complexes exhibited many sharp peaks, all of
which can be assigned to Y zeolite. This suggests that
the framework of the zeolite was kept intact even after
the heat treatment. The XRD patterns for the PAN and

A New Type of Porous Carbon

Chem. Mater., Vol. 9, No. 2, 1997 613

Figure 5. XRD patterns for the CVD carbons prepared in the

USY(12) zeolite under different carbon deposition periods.

width of each (002) peak. The values of d002 were 0.35,

0.35, and 0.36 nm for the PAN carbons from NaY(5.5)
and NaY(5.6) and the PFA carbon from NaY(5.6),
respectively. These values are much larger than that
of ideal graphite (d002 ) 0.3354 nm), indicating low
crystallinity of these carbons. The Lc values for the
polymer carbons ranged from 1.4 to 1.5 nm.

Figure 3. TEM images of the CVD carbons prepared in the

USY(12) zeolite under different carbon deposition periods (a
microscope grid is seen under the samples).

Figure 4. XRD patterns for the polymer carbons prepared

in NaY zeolite.

PFA carbons are given in Figure 4, where a broad (002)

diffraction peak is observed, but no other peak appears.
The structural parameters such as the interplanar
spacing (d002) and the average crystallite size (Lc) were,
respectively, determined from the position and the half-

Figure 5 illustrated the XRD patterns of the CVD

carbons with different carbon deposition periods. The
pattern depends on the carbon deposition period. In the
case of the sample with the 3 h deposition, we can
observe no peak at around 25. After a 6 h deposition,
the (002) peak is still unclear. The 12 h carbon
deposition, however, gives a relatively sharp (002) peak
with d002 of 0.351 nm and Lc of 2.8 nm.
N2 Adsorption. Figure 6 shows the N2 isotherms
on the zeolites and carbon/zeolite complexes at -196 C.
As well as the zeolite samples, the complexes show a
rapid increase in their adsorption isotherms at a very
low relative pressure. Even for the complex prepared
by the 6 h carbon deposition, there is a small increase
in adsorption at a low pressure (Figure 6i). This finding
indicates that these complexes still have micropores. In
the cases of the complex with the 12 h deposition (not
shown here), its isotherm did not show such a rapid
increase. This may be due to the pore blockage on the
external surface of the zeolite particles. We will discuss
more details of this later.
The N2 isotherms on the carbon samples are illustrated in Figure 7, where all the isotherms suggest
the presence of micropores in these carbons. This figure
clearly indicates that N2 adsorbs on the CVD carbons
much more than on the PAN and PFA carbons. From
these isotherms, we determined their BET surface area,
micropore and mesopore volumes, which are summarized in Tables 2 and 3. Table 3 also includes the
data of the CVD carbons prepared under the other
deposition conditions. The values of the surface area
and the pore volumes for the CVD carbons are significantly larger than those for the polymer carbons. Some
of the CVD carbons have BET surface areas of more

614

Chem. Mater., Vol. 9, No. 2, 1997

Kyotani et al.
Table 3. Effect of Carbon Deposition Conditions on Pore
Structure of the CVD Carbons Prepared in the Zeolite
Channels
propylene
BET
deposition
concn
deposition surface
micropore mesopore
temp (C) (% in N2) period (h) area (m2/g) vol (cm3/g) vol (cm3/g)
800
800
800
800

2.5
2.5
2.5
2.5

1
3
6
12

1320
1720
1510
1430

0.54
0.69
0.55
0.57

0.42
0.71
0.64
0.52

800
800
800

1.3
1.3
1.3

12
15
18

2060
2200
1790

0.82
0.88
0.72

0.75
0.83
0.62

700
700

2.5
2.5

12
18

1660
2260

0.66
1.11

0.79
0.76

ene concentration (1.3%) favors the production of high

surface area carbon, that is, high surface area can be
obtained at a low carbon deposition rate. Too long
carbon deposition, however, reduces its surface area.
Discussion

Figure 6. N2 isotherms for the zeolites and the carbon/zeolite

complexes: (a) NaY(5.5); (b) NaY(5.6); (c) PAN carbon/NaY(5.5); (d) PAN carbon/NaY(5.6); (e) PFA carbon/NaY(5.6); (f)
USY(12); (g), (h), and (i) CVD carbon/USY(12) complexes
prepared under carbon deposition periods of 1, 3, and 6 h,
respectively. For each isotherm, the upper and lower curves
correspond to the desorption and adsorption isotherms, respectively.

Figure 7. Adsorption isotherms for N2 on the carbons

prepared in the zeolite channels: (a) PAN carbon from NaY(5.5); (b) PAN carbon from NaY(5.6); (c) PFA carbon from NaY(5.6); (d), (e), (f), and (g) CVD carbons from USY(12) under
carbon deposition periods of 1, 3, 6, and 12 h, respectively.
Table 2. Pore Structure of the Polymer Carbons
Prepared in the Zeolite Channels
sample
PAN carbon from
NaY(5.5)
NaY(5.6)
PFA carbon from
NaY(5.6)

BET surface
area (m2/g)

micropore
vol (cm3/g)

mesopore
vol (cm3/g)

700
580

0.31
0.28

0.20
0.10

590

0.28

0.15

than 2000 m2/g. The micropore volume of each polymer

carbons is always larger than its mesopore volume,
while the micro- and mesopores have comparable volume for most of the CVD carbons.
We can evaluate from Table 3 the optimum deposition
conditions for obtaining carbon with high surface area.
A lower reaction temperature (700 C) or lower propyl-

Table 1 indicates that most of the carbon/zeolite

complexes include about 0.2 g of carbon/g of zeolite. If
these large amounts of carbons are present mainly on
the external surface of the zeolite crystals, we should
have observed significant difference in morphology
between the pure zeolite crystals and the corresponding
complexes. Moreover, their TEM images should have
shown a void inside each carbon particles, because the
removal of zeolite framework mainly leaves carbon
envelope over the particles. The SEM and TEM observations, however, did not give such microscopic features;
we observed no remarkable carbon deposit on the
external surface of the zeolite crystals and no clear void
in each particle. This is more obvious with the TEM
image of the sliced carbon samples (Figure 2b), where
we observed no void inside each particle. This is a direct
evidence of the presence of carbon in the nanochannels
of the zeolites.
According to the Dubinin-Radushkevich plot, the
micropore volume of the zeolites was calculated to be
about 0.30 cm3/g for USY(12), 0.38 for NaY(5.5), and
0.36 for NaY(5.6). If the carbon is packed only in the
micropores of the zeolite, the maximum amount of
packed carbon can be calculated by assuming that the
carbon in the nanochannels has the same density as
that of graphite (2.26 g/cm3). These values were 0.68,
0.86, and 0.81 g for USY(12), NaY(5.5), and NaY(5.6),
respectively. The actual amounts of carbon are always
less than the theoretical one (see Table 1). Furthermore, the isotherms of the complexes suggest that some
micropores are still present in the complexes. Hence,
in every case the carbon packing does not reach complete filling in the zeolite channels. In the case of the
polymer carbon complexes, such incomplete filling would
be mainly due to the weight loss of the polymer during
its carbonization in the channels.
The CVD carbon with a 12 h deposition has dense
carbon layers (Figure 3). Such carbon deposition would
result in the complete pore blockage of zeolite, which
explains the absence of micropores in the corresponding
carbon/zeolite complex.
Despite the fact that the polymer carbons gave the
carbon (002) diffraction peak, the XRD patterns for the
corresponding carbon/zeolite complexes did not exhibit
this carbon peak. One of the possible reasons for this
is that the amount of the polymer carbon in the

A New Type of Porous Carbon

complexes was not large enough to give a (002) peak.

To clarify this point, we made an XRD analysis for a
physical mixture of NaY(5.6) and the PAN carbon
isolated from a NaY(5.6) matrix. The amount of carbon
in this mixture was adjusted to be equal to that in
thePAN carbon/NaY(5.6) complex. The resultant XRD
pattern of this mixture showed a very weak but detectable (002) diffraction. The absence of the (002) peak in
the complexes must be, therefore, ascribed not to the
low carbon content but to another reason. The most
probable explanation is that a carbon stacking structure
cannot be developed in such a small space as a zeolite
channel. For the polymer carbons, the appearance of
(002) peak after the acid washing suggests that carbon
layers were stacked together upon the removal of zeolite
framework. The Lc values for the polymer carbons (1.41.5 nm) are larger than the size of a supercage (1.2 nm)
of Y zeolite. This fact also suggests that carbon layers
were stacked after they were extracted from the zeolite.
From the SEM photographs of these carbons, we observed the size of the carbon particles to be always
smaller than that of the corresponding carbon/zeolite
complex particles, despite little morphological change.
The carbon layer stacking during the washing might be
related to this shrinkage in size.
In contrast to the carbons from the polymers, the CVD
carbon with the 3 h deposition did not give any (002)
peak even after isolation, indicating no stacking structure and extremely poor crystallinity of this carbon. A
similar XRD pattern was reported for the activated
carbons with high surface area, which were activated
by KOH.16 On the other hand, a sharp (002) peak was
observed for the carbon with the 12 h deposition. The
possible origin for this sharp peak would be the dense
carbon layers on the outer surface of the carbon particles
as observed in Figure 3.
The N2 adsorption results revealed that all the
carbons prepared in the zeolite channels are highly
porous. The porosity of these carbons originates from
the porous structure of the Y zeolite. The zeolite
framework in the carbon/zeolite complex would convert
to the pore of the resultant carbon after the acid
washing. The CVD carbons have a higher surface area,
and some of them are more than 2000 m2/g. The lower
surface area and pore volumes of the polymer carbons
can be partially attributed to the stacking of carbon
layers during the acid washing, because such stacking
(16) Shiraishi, M.; Yoshizawa, N.; Ohkawa, K.; Maeda, Tanso 1992,
155, 295.
(17) Marsh, H.; Crawford, D.; OGrady, T. M.; Wennerberg, A.
Carbon 1982, 20, 419.
(18) Matsumura, Y. Chem. Chem. Ind. 1990, 43, 358.
(19) Maeda, T. In Kyuchaku Gijyutsu Handbook; Shimizu, H., Ed.;
NTS: Tokyo, 1993; Chapter II-1-3; p 154.

Chem. Mater., Vol. 9, No. 2, 1997 615

may cause the disappearance of some pores and/or the

reduction in pore size.
To the best of our knowledge, this study is the first
report on preparing porous carbon with a surface area
of more than 2000 m2/g without any conventional
activation using gas or chemicals. Furthermore, our
method makes it possible to prepare high surface area
carbon without carbon consumption, while any activation process consumes carbon to a substantial extent.
For example, activated carbons prepared by chemical
treatment with KOH have a high BET surface area of
2500-4500 m2/g,16-18 but such a high surface area as
4500 m2/g was obtained with a carbon loss of as much
as 75%.19 Most of the pores in this type of activated
carbon are dominated by micropores.18,19 The porosity
of the CVD carbons prepared by the present method,
however, consists of not only micropores but also a
considerable amount of mesopores.
Conclusions
For the carbon/zeolite complexes, we confirmed the
presence of carbon in the nanochannels of Y zeolite. The
carbon packing in the channels was, however, not
complete, and there are still some space unoccupied by
carbon. The carbon in the channels did not have any
regular stacking structure as carbon usually has. This
is due to the spatial limitation of the zeolite channels.
It was found that the microscopic morphology of each
zeolite exactly reflects that of the corresponding carbon.
All of these carbons are highly porous. We can conclude
that the channel structure of Y zeolite endowed the
resultant carbon with its porosity. The CVD carbons
have higher surface area and larger pore volume than
those of the polymer carbons. The BET surface area of
the former carbons could reach more than 2000 m2/g,
when their deposition conditions were optimized. This
is the first example of preparing porous carbon with
such a high surface area without any activation process.
Unlike the high surface area carbon prepared by chemical activation with KOH, mesopores, as well as micropores, are well developed in the CVD carbons.
Acknowledgment. We thank the High Voltage
Electron Microscope Laboratory of Tohoku University
and Dr. N. Sonobe of Kureha Chemical Industry Co.,
Ltd., for microscopical analysis. We are indebted to Dr.
H. Terasaki of Tohoku University for the preparation
of zeolite sample. Thanks are also due to Dr. J.
Rodrguez-Mirasol of University of Malaga for his
helpful discussion. This study was partly supported by
Special Coordination Funds for Promising Science and
Technology from Science and Technology Agency, Japan.
CM960430H